Error Intelligence
Pattern Clustering
Updated Oct 27, 2026 · today
Groups errors with similar templates so many variants of the same problem cluster into one alert. Surfaces brand-new clusters as they appear.
Example alert
Exact wording varies — the detector generates titles from the anomaly it finds. This is representative of what an alert looks like when it fires.
How it works
Template mining extracts the structural shape of each error message, reducing thousands of unique strings to a small set of recurring patterns. Semantic similarity then merges near-duplicates across services. New clusters surface the moment they first appear.
Availability
Runs on these tiers:
Want to see this detector firing in the live demo?
Open alerts in the sandbox →Related detectors
- New Error Detection
Catches errors that have never appeared in your 7-day baseline. On connect, the baseline seeds from your last 7 days of historical logs — push history for day-1 alerts, or wait a week for organic warm-up.
- Error Concentration
Raises warnings when elevated service errors remain concentrated on a field value compared with the baseline.
- New Frontend Error
Raises warnings for recurring browser error patterns absent from the recent baseline.
- New Mobile Crash
Raises warnings for new crash patterns across multiple sessions in an app, with the early-access mobile SDK.